Upload
others
View
4
Download
0
Embed Size (px)
Citation preview
U.S. Department of the Interior
U.S. Geological Survey
Futurecasting effects of sea
level rise, climate change, and
restoration on individual species
Brad Stith, Zuzanna Zajac, Catherine A.
Langtimm, Eric D. Swain, Don DeAngelis, and
Melinda Lohmann
Hydrodynamic/Climate Models
Abundance of futurecast data from
hydrology/climate models becoming
available.
Huge data volume creates opportunity and
data analysis problem.
Need to incorporate futurecast data into
biological models.
Long temporal scale and large spatial extent
dictate use of simple biological models.
Bisect
TEN THOUSAND
ISLANDS
FLORID
A
MODEL STUDYAREA
24°30́
25°00́
25°30́
26°00́
26°30́
27°00́83°00´ 82°30´ 82°00́ 81°30́ 81°00́ 80°30́ 80°00́
S. Florida Hydrodynamic Models
Before Picayune Strand Restoration Project
After Picayune Strand Restoration Project
Sample output (Figures) Ten-Thousand Islands salinity before and after Picayune Strand Restoration Snapshot: 01 Oct. 2003
Hydrology Model Output • Salinity • Temperature • Stage/depth
Scenarios • CERP restorations • Sea Level Rise
Resolution • 15 minute time step • 500 meter grid cell
Biological Models
Need a simple approach to compare biological implications of different scenarios of restoration, sea level rise, and climate change
Habitat Suitability Index (HSI) and Spatially Explicit Species Index (SESI) models do not require extensive biological datasets
incorporate spatial and temporal variation
allow relative comparisons of different scenarios
model potential habitat suitability, not predicting occurrence
Biological Research Focus
Submerged Aquatic
Vegetation (SAV)
Vallisneria americana (Tape
grass) – freshwater species
Halodule wrightii (Shoal grass)
– salt-tolerant species
Florida Manatee
HSI Model H. wrightii_salinity
Salinity [-]
0 10 20 30 40 50 60 70
HS
I Sa
lin
ity
0.0
0.2
0.4
0.6
0.8
1.0
H. wrightii_light
ADBL [micro mol m-2 s-1]
0 100 200 300 400 500
HS
I AD
BL
0.0
0.2
0.4
0.6
0.8
1.0
H. wrightii_temperature
Temperature [C]
10 15 20 25 30 35 40 45 50
HS
I Te
mp
era
ture
0.0
0.2
0.4
0.6
0.8
1.0
V. americana_salinity
Salinity [-]
0 10 20 30 40 50 60 70
HS
I Sa
lin
ity
0.0
0.2
0.4
0.6
0.8
1.0
V. americana_temperature
Temperature [C]
10 15 20 25 30 35 40 45 50
HS
I Te
mp
era
ture
0.0
0.2
0.4
0.6
0.8
1.0
V. americana_light
0 100 200 300 400 500
HS
I AD
BL
0.0
0.2
0.4
0.6
0.8
1.0
ADBL1
ADBL2
ADBL [micro mol m-2 s-1]
Halodule wrightii Vallisneria americana
Salinity
Temperature
Light/depth
Habitat Suitability Indices (HSIs)
3Total Salinity Temperature ADBLHSI HSI HSI HSI
• Calculated for each grid cell
and every time step
• HSITotal for a cell depends only
on environmental variables in
each cell (i.e. is independent
from neighboring cell values).
Global Uncertainty/Sensitivity Analysis Steps
Uncertainty
Analysis (UA)
factor3
factor2
factor1
RG
LOB
ALL
Y SA
MP
LED
INP
UT
FAC
TOR
S
MO
DEL
OU
TPU
TS
Global Sensitivity
Analysis (GSA)
0%
20%
40%
60%
80%
100%
0
50
100
150
200
250
300
350
0. 00 0. 01 0. 01 0. 02 0. 02 0. 03 0. 03 0. 04 0. 04 0. 05 0. 05 0. 06 0. 06 0. 07 0. 07 0. 08 0. 08 0. 09 0. 09 0. 10 0. 10 0. 11 0. 11 0. 12 0. 13 0. 13 0. 14 0. 14 0. 15 0. 15 0. 16 0. 16 0. 17 0. 17 0. 18 0. 18 0. 19 0. 19 0. 20 0. 20 0. 21 0. 21 0. 22 0. 22 0. 23 0. 23 0. 24 0. 25 0. 25 0. 26 0. 26 0. 27 0. 27 0. 28 0. 28 0. 29 0. 29 0. 30 0. 30 0. 31 0. 31 0. 32 0. 32 0. 33 0. 33 0. 34 0. 34 0. 35 0. 36 0. 36 0. 37 0. 37 0. 38 0. 38 0. 39 0. 39 0. 40 Mor eBin
HSI
MODEL
3
6
5
2 1
4
1. Identify uncertain spatially distributed inputs and define uncertainty models
(PDFs).
2. Generate input values pseudo randomly from assigned PDFs using Sobol
method.
3. Run the model for multiple alternative input sample (Monte Carlo).
4. Construct PDF for the model output (from N output values).
5. Perform SA using SIMLAB.
Specification of uncertainty for look-up tables, using
HSIsalinity vs. salinity lookup table as an example variable
0
0.2
0.4
0.6
0.8
1
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70
HS
I_S
ali
nit
y
Salinity [-] Uniform ± 20% around base value
Uniform ± 20% around base value,
truncated to 1 for the upper limit
Mean HSI (left) and Uncertainty/SD (right)
for Vallisneria americana.
Benchmark Cells for Sensitivity Analysis
1
2
3
4
5 6
7 8 9
10
1. Area 1 (Turner River) 2. Area 2 (Sunday Bay Upper) 3. Area 5 (Alligator Bay) 4. Area 6 (Lostmans Second Bay) 5. Area 10 (Broad River Bay) 6. Area 11 (Upper Broad River) 7. Area 12 (Harney River) 8. Area 13 (Tarpon Bay) 9. Area 14 (Upper Shark River) 10. Area 16 (Lower Shark River)
Shark River
Dry Season Salinity Trends Wet Season Salinity Trends
Mean Salinity (5-year mean)
Salinity Variance
PDFs of HSI values – 4 ENP sites
H_2_6
HSITotal
0.0 0.2 0.4 0.6 0.8 1.0
Pro
bability
0.0
0.1
0.30.50.81.0
cell2 cell5 cell8
V_2_6
HSITotal
0.0 0.2 0.4 0.6 0.8 1.0
Pro
bability
0.0
0.1
0.30.50.81.0
cell 9
Halodule wrightii Vallisneria americana
Habitat suitability PDFs reflect uncertainty,
but show high and low suitabilities for the 2
SAV species that differ among sites
Sensitivity Indices – 4 ENP sites
Vallisneria americana Halodule wrightii
Halodule model shows more sensitivity to light,
Vallisneria model shows more sensitivity to salinity
Picayune Strand Restoration (lower left map) shows reduced bay salinities, but differences absent with sea level rise (July 1998-2008 mean)
TTI Salinity Difference Maps
Fakahatchee Bay salinity differences for 6 scenarios
Picayune Strand Restoration shows reduced salinities compared to sea level rise and existing condition scenarios. Variance peaks during beginning and end of wet season.
Salinity (mean) Variance
Fakahatchee Bay Halodule HSI differences for 6 scenarios
Habitat suitability for Halodule is lower for sea level rise scenarios at this site. Variance is high, especially at dry-wet season transition.
HSI (mean) HSI Uncertainty
Summary HSI/SESI approach provides a simple modeling
framework to analyze and compare biological implications of large futurecast datasets and alternative restoration scenarios
Uncertainty and Sensitivity Analysis shows which model parameters produce the greatest variation and provide estimates of model uncertainty in space and time can help direct monitoring resources to measure
parameters and sites with greatest uncertainty and sensitivity
uncertainty maps can help managers evaluate model results
Difference maps and graphs of changes in habitat suitability can reveal trends and relative differences associated with restoration and sea level rise.
Acknowledgements
USGS FISCHS (Future Impacts of Sea
Level Rise on Coastal Habitats and
Species)
Funding from USGS Greater Everglades Priority Ecosystems Science
USGS Ecosystems Mapping
USGS Global Change Research & Development
Program